Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

01.
arXiv (CS.LG) 2026-06-17

When the Next Step Is Not One Step: Distribution-Aware Execution Modeling for Concurrent Go Programs

arXiv:2606.17508v1 Announce Type: new Abstract: Training a model to predict the next step in a concurrent program is harder than it looks: two runs of the same program from the same trace prefix can produce different next events, both valid, because the scheduler is nondeterministic. A model trained against a single label is learning to guess one outcome of a random process. We turn this around and use the nondeterminism as a training signal. We run each program many times, aggregate the observed next events into an empirical distribution, and fine-tune a 7B model to match that distribution with a KL objective. On 798 held-out predictions drawn from real production Go bugs (CockroachDB, Kubernetes, gRPC, etcd), fine-tuning on fewer than a thousand traces reaches 36.2% accuracy, ahead of Gemini 3.5 Flash used zero-shot (34.8%) and the same model without fine-tuning (28.6%). Distribution training matches cross-entropy on accuracy (35.8% vs. 36.2%) while reducing Expected Calibration Error from 0.205 to 0.169. We also derive a formal goroutine-leak signature for a class of select-blocked goroutines where P(GoUnblock)=0 holds by scheduler semantics, not by learning. We release the dataset, trained adapters, and all tooling.

02.
arXiv (CS.CL) 2026-06-12

LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories

Scientific laboratories increasingly rely on AI systems to reason about experiments, but the physical act of doing science remains largely outside their reach. AI can help read literature, generate hypotheses, and plan protocols, yet the execution of those protocols at the bench still requires a human operator. Vision-Language-Action (VLA) models provide one possible interface between written protocols and robot execution, but existing policies are trained mostly on household and tabletop demonstrations and rarely encounter the instruments, transparent liquids, or fixed protocol workflows found in scientific laboratories. Closing this gap requires both laboratory-specific supervision and a unified learning framework that can accommodate the diverse robot embodiments used to execute experimental protocols. We therefore identify data and embodiment as central bottlenecks alongside model design. To address the data side, we build RoboGenesis, a simulation-based workflow and data engine that composes configured laboratory workflows from atomic skills, validates and filters rollouts, and exports structured demonstrations across supported robot profiles. On the policy side, we present LabVLA, trained with a two-stage recipe: FAST action token pretraining first makes the Qwen3-VL-4B-Instruct backbone action aware before any continuous control is learned, and flow matching posttraining then attaches a DiT action expert under knowledge insulation. On the LabUtopia benchmark, LabVLA achieves the highest average success rate among all evaluated baselines under both in-distribution and out-of-distribution settings.

03.
arXiv (CS.CL) 2026-06-19

Analyzing Error Propagation in Korean Spoken QA with ASR-LLM Cascades

We analyze how automatic speech recognition (ASR) errors propagate through ASR-LLM cascades in Korean spoken question answering (SQA), focusing on downstream semantic failures that conventional ASR metrics cannot fully capture. Our analysis shows that the relative downstream degradation caused by ASR errors is consistent across LLMs with different absolute performance, suggesting that cascade degradation largely tracks ASR-stage information loss. We further identify single-character Korean ASR errors as a Korean-specific loss channel, where even a minimal transcription difference can change the intended question and degrade downstream QA performance. Finally, an auxiliary comparison shows that a large audio language model outperforms an ASR-LLM cascade with an approximately matched language backbone in noisy Korean SQA, indicating the potential of direct audio input to mitigate transcript-induced information loss.

04.
arXiv (math.PR) 2026-06-18

Power Partitions and Hayman Functions

arXiv:2602.18575v3 Announce Type: replace Abstract: We prove, within the probabilistic framework of Khinchin families, that the generating function $P_k$ of partitions into $k$-th powers is strongly Gaussian in the sense of Báez-Duarte, and even further that it is a Hayman function. Thus the Hardy–Ramanujan asymptotic formula for the number $p_k(n)$ of partitions of $n$ into $k$-th powers which reads \[ p_k(n) \sim \frac{\alpha_k}{n^{(3k+1)/(2k+2)}} \exp\!\Big(\beta_k\, n^{1/(k+1)}\Big), \qquad n\to\infty, \] where $\alpha_k$ and~$\beta_k$ are explicit constants depending only on $k$, follows directly from Hayman's asymptotic formula for strongly Gaussian power series. The proof of strong Gaussianity of $P_k$ combines a Gaussianity criterion for Khinchin families with certain bounds of Tenenbaum, Wu and Li on the generating function; the asymptotic formula is recovered by computing asymptotic approximations of the mean and variance of the associated family. Analogous results are presented for the generating function $Q_k$ of partitions into distinct $k$-th powers.

05.
arXiv (CS.AI) 2026-06-16

Combining Retrieval-Augmented Text Generation with LLMs for Reading Content Recommendations

arXiv:2606.14817v1 Announce Type: cross Abstract: This work presents the design, implementation, and evaluation of a system for generating personalized reading content using Large Language Models (LLMs) combined with Retrieval-Augmented Generation (RAG). The proposed architecture consists of four modules: Input, RAG, Generation, and Judging and enables users to specify both a question and a target reading content complexity. RAG is employed to retrieve relevant information from the Internet, enriching and grounding the content produced by three modern LLMs: Meta LLaMA 4 Scout, LLaMA 3.1 8B Instant, and Google Gemma2 9B. Reading materials are generated using three prompting strategies (Chain-of-Thought, zero-shot, and few-shot), and the LLM-as-a-Judge module automatically evaluates answer quality and alignment with the desired readability level. Experimental results show that RAG consistently improves system performance across all models and prompting techniques, increasing relevance and particularly groundedness by up to 26-35 percentage points. Overall, the findings demonstrate that the RAG-augmented architecture effectively produces reading content tailored to user queries and desired textual complexity.

06.
arXiv (CS.CV) 2026-06-17

Flux-Guard: Facial Identity Protection using diffusion models

The widespread deployment of face recognition (FR) systems exposes personal images shared on social media and public platforms to identity linkage and privacy risks. Existing adversarial privacy protection methods can degrade unauthorized FR performance but are not compatible with generative face editing. Artificial intelligence-driven face editing tools are gaining popularity, which has significantly increased user demand for personalized portrait generation and social sharing. However, current editing methods often preserve identity features, making the edited images still susceptible to tracking by malicious FR systems. Thus, this paper proposes Flux-Guard, a privacy-preserving face editing framework based on adversarial attacks, which integrates face editing and privacy protection within a unified generative process. Specifically, we design a flow trajectory control method to align semantic manipulations with the generative process and introduce latent-space adversarial optimization with an adaptive perceptual-loss-driven weighting strategy, dynamically adjusting adversarial strength to maximize attack effectiveness while preserving visual quality. Extensive experiments demonstrate that Flux-Guard supports face editing while significantly improving attack success rates against cross-domain face recognition models on the CelebA-HQ and LADN datasets. Furthermore, evaluation results for commercial APIs have confirmed its effectiveness in real-world applications. The code is released at https://github.com/JLMWang/Flux-Guard.

07.
arXiv (CS.CV) 2026-06-15

Explaining RhythmFormer: A Systematic XAI Analysis of Periodic Sparse Attention for Remote Photoplethysmography

Remote photoplethysmography (rPPG) transformers achieve low heart-rate error on benchmarks, yet their decisions remain opaque–a growing concern as rPPG moves toward clinical heart rate estimation. Existing rPPG XAI is dominated by qualitative heatmap inspection without quantitative faithfulness metrics or physiology-grounded validation, leaving a gap between visual plausibility and auditable evidence. We address this gap. First, we adapt four attribution methods (raw attention, rollout, flow, Beyond Intuition) to RhythmFormer's bi-level routing attention with top-$k$ selection. Second, we introduce a skin coverage metric quantifying how much attribution mass falls on skin regions. Third, we adapt the SaCo faithfulness coefficient from its original classification setting to rPPG regression by using the MAE between original and perturbed predicted rPPG waveforms as the perturbation impact. Applying these tools, we quantify a multi-hop leakage effect under sparse top-$k$ routing: attention rollout and flow almost completely restores the connections that individual refined-attention layers explicitly set to zero. Beyond Intuition mitigates this via its value-projection-weighted rollout and gradient-supported mask, attaining the highest median refined skin coverage ($0.83$ vs. $0.57$ for vanilla rollout) and faithfulness ($F=0.92$) among the evaluated methods on UBFC-rPPG. Validation across diverse datasets and model variants is needed. A case study on a low-SaCo outlier further shows all four methods recovering consistently once an artefactual region is replaced, suggesting consistent SaCo behavior across attribution families in this illustrative case. Together, these metrics move XAI for rPPG toward auditable numerical evidence about spatial alignment and perturbation faithfulness, i.e. trustworthy rPPG XAI.

08.
arXiv (CS.CL) 2026-06-17

ART: Attention Run-time Termination for Efficient Large Language Model Decoding

Long-context decoding in Large Language Models (LLMs) is constrained by the cost of accessing and processing the Key-Value (KV) cache. Despite evidence that attention outputs depend jointly on keys and values, most existing KV management methods rely on key-only pruning, since incorporating values incurs prohibitive overhead. In this paper, we propose Attention Run-time Termination (ART), a lightweight run-time mechanism that tracks accumulated attention outputs during kernel execution and terminates subsequent KV block accesses once further contributions become negligible. Rather than replacing KV selection, ART dynamically terminates redundant KV traversal on top of existing dense or sparse attention policies. We introduce a stability-based criterion that monitors both magnitude and directional changes of intermediate attention outputs and provideds a theoretical characterization of the resulting truncation error. Experiments on the LongBench and RULER Needle-in-a-Haystack tasks show that ART increases the generation throughput of existing KV-cache methods by up to 20%, without compromising the result quality.

09.
arXiv (CS.LG) 2026-06-16

Finite-Width Neural Tangent Kernels from Feynman Diagrams

arXiv:2508.11522v4 Announce Type: replace Abstract: Neural tangent kernels (NTKs) are a powerful tool for analyzing deep, non-linear neural networks. In the infinite-width limit, NTKs can easily be computed for most common architectures, yielding full analytic control over the training dynamics. However, at infinite width, important properties of training such as NTK evolution or feature learning are absent. Nevertheless, finite width effects can be included by computing corrections to the Gaussian statistics at infinite width. We introduce Feynman diagrams for computing finite-width corrections to NTK statistics. These dramatically simplify the necessary algebraic manipulations and enable the computation of layer-wise recursion relations for arbitrary statistics involving preactivations, NTKs and certain higher-derivative tensors (dNTK and ddNTK) required to predict the training dynamics at leading order. We demonstrate the feasibility of our framework by extending stability results for deep networks from preactivations to NTKs and proving the absence of finite-width corrections for scale-invariant nonlinearities such as ReLU on the diagonal of the Gram matrix of the NTK. We numerically implement the complete set of equations necessary to compute the first-order corrections for arbitrary inputs and demonstrate that the results follow the statistics of sampled neural networks for widths $n\gtrsim 20$.

10.
arXiv (CS.LG) 2026-06-19

Matching Markets meet Cumulative Prospect Theory: Towards Optimal and Adversarially Robust Learning

arXiv:2606.19883v1 Announce Type: new Abstract: We study a multi-agent multi-armed bandit problem in the competitive setup with two-sided matching markets under a human centric decision making model. To capture human preferences, we use cumulative prospect theory (CPT) that weighs the actions of the agent in a nonlinear fashion using a ($\alpha$-Hölder continuous) weight function. CPT has been widely used in behavioral economics and risk sensitive machine learning to emulate human preferences. We analyze the state-of-the-art learning algorithm with CPT weight distorted rewards and obtain a player optimal regret of $\mathcal{O}(K\log T \left(\frac{1}{\Delta}\right)^{2/\alpha})$, where $K$ denotes the number of arms, $T$ is the learning horizon, and $\Delta$ represents (suitably defined) players' minimum preference gap. Noticing the dependence on $\Delta$ to be sub-optimal, we further improve this regret by judiciously selecting the active set of arms during exploration, which removes the dependence on $K$ in the dominant term and achieves an improved (optimal) regret guarantees in the setting where the number of arms $K$ is significantly larger than the number of players $N$. In addition, we consider adversarial markets where the observed rewards of the agents may be corrupted. We propose and analyze algorithms for robust markets with CPT as risk sensitive measure in both settings where the total corruption budget is known and where it is unknown, and establish logarithmic player-optimal regret guarantees in both cases.

11.
arXiv (CS.CV) 2026-06-18

Confidence is Not Reliability: Rethinking MC Dropout in Brain Tumour Segmentation

Glioma segmentation in multiparametric MRI is a critical component of treatment planning. A segmentation model that fails silently on treatment-critical sub-regions represents a patient safety risk that overlap-based metrics such as Dice scores cannot expose. We ask whether voxel-level uncertainty estimation via Monte Carlo (MC) Dropout can reliably identify segmentation errors in clinically critical sub-regions, and whether calibration failure modes are detectable from standard reporting metrics alone. In an empirical two-model case study on 126 BraTS21 patients, we evaluate a high-performance pretrained SegResNet and a locally trained UNet with residual units (UNet-Res). MC dropout preserved segmentation accuracy ($|\Delta Dice|$ $

12.
arXiv (CS.AI) 2026-06-12

SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models

arXiv:2602.04208v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models have emerged as a promising paradigm for general-purpose robotic control, with test-time scaling (TTS) gaining attention to enhance robustness beyond training. However, existing TTS methods for VLAs require additional training, verifiers, and multiple forward passes, making them impractical for deployment. Moreover, they intervene only at action decoding while keeping visual representations fixed-insufficient under perceptual ambiguity, where reconsidering how to perceive is as important as deciding what to do. To address these limitations, we propose SCALE, a simple inference strategy that jointly modulates visual perception and action based on 'self-uncertainty', inspired by uncertainty-driven exploration in Active Inference theory-requiring no additional training, no verifier, and only a single forward pass. SCALE broadens exploration in both perception and action under high uncertainty, while focusing on exploitation when confident-enabling adaptive execution across varying conditions. Experiments on simulated and real-world benchmarks demonstrate that SCALE improves state-of-the-art VLAs and outperforms existing TTS methods while maintaining single-pass efficiency.

13.
arXiv (CS.AI) 2026-06-17

MemTrace: Probing What Final Accuracy Misses in Long-Term Memory

arXiv:2606.17328v1 Announce Type: new Abstract: LLM agents increasingly maintain long-term memory of user facts across sessions. Yet such memory is usually evaluated by aggregating accuracy over question rows or episodes. Because this approach scores question rows independently, even when several questions probe the same fact, it cannot show how that fact behaves as conditions change. We introduce MemTrace, a benchmark whose unit of measurement is the knowledge point: a single typed fact about the user, rather than an individual question. MemTrace probes each fact along three controlled dimensions: memory age, defined by how many sessions ago the fact appeared in the history; question type, covering current state, earlier state, and trajectory of change; and evidence condition, covering present, missing, and contradicted-by-false-premise settings. Evaluating 13 memory-system configurations across four paradigms, we find that similar pooled accuracy hides different failures: recovering a fact's current and earlier states does not imply tracking how it changed, and safe abstention does not imply correcting a false premise. The dominant bottleneck is evidence use, not retrieval: when systems fail, the evidence was retrievable 10 times more often than it was missing. These results suggest that improving long-term memory requires better use of reachable evidence, not simply more storage or retrieval.

14.
arXiv (CS.AI) 2026-06-16

Infant Spontaneous Movement Noise Improves Exploration in Deep RL

arXiv:2606.16590v1 Announce Type: cross Abstract: Exploration in deep reinforcement learning (RL) is commonly implemented as temporally uncorrelated white noise. However, recent works show that temporally correlated colored noise can improve exploration efficiency by producing smooth trajectories with better coverage of the state space. We inquire whether action noise inspired by infant spontaneous movements can also improve exploration in deep RL. We find that the power spectral densities of babies' end-effector velocities follow a colored noise process where the spectral exponent increases with age. Inspired by this developmental pattern, we introduce a mechanism that progressively increases the temporal auto-correlation of exploration noise during RL training, matching the infant statistics. Experiments across several RL environments show that infant-inspired noise produces structured exploratory behavior and can improve learning efficiency compared to conventional exploration strategies. These findings suggest that human motor and cognitive development can provide useful guidance for designing learning mechanisms in artificial agents. Our code is available at https://github.com/trieschlab/baby-noise-rl.

15.
arXiv (CS.LG) 2026-06-18

FORGE: Foundational Optimization Representations from Graph Embeddings

arXiv:2508.20330v5 Announce Type: replace Abstract: Combinatorial optimization problems are ubiquitous in science and engineering. Still, learning-based approaches to accelerate combinatorial optimization often require solving a large number of difficult instances to collect training data, incurring significant computational cost. Existing learning-based methods require training dedicated models for each problem distribution, for each downstream task, severely limiting their scalability and generalization. We introduce Forge: Foundational Optimization Representations from Graph Embeddings, a framework that pre-trains a vector-quantized graph autoencoder on a large, diverse collection of mixed-integer programming (MIP) instances in an unsupervised manner, without relying on optimization solvers or optimal solutions. Vector quantization produces discrete code assignments that serve as a vocabulary for representing optimization instances. We evaluate Forge in both unsupervised and supervised settings. In the unsupervised setting, Forge embeddings effectively cluster unseen instances across problem domains and sizes. In the supervised setting, we fine-tune Forge embeddings and show that a single pre-trained model helps predicting both the integrality gap for cut-generation and variable hints for search guidance across multiple problem and size distributions. In both tasks, we improve the performance of a commercial optimization solver and outperform state-of-the-art learning-based methods. Finally, we open-source our training code, pre-trained Forge weights, and embeddings for multiple MIP distributions to foster further research in representation learning for optimization problems https://skadio.github.io/forge/

16.
arXiv (CS.AI) 2026-06-16

MiroBench: Benchmarking Realism in Agentic Simulation of Real-world Discussions

arXiv:2606.14715v1 Announce Type: cross Abstract: LLM agents are increasingly used to simulate real world interactions, but it remains unclear whether simulated behaviors preserve the content patterns and interaction dynamics of real human behaviors. Existing evaluations remain fragmented, which makes it difficult to compare systems or measure progress. In this paper, we focus on Reddit discussions as a concrete first step toward evaluating real-world social simulation. Reddit threads provide public, topic-grounded, multi-party interactions where people share experiences, debate, seek advice, express emotion, and collectively respond to products, events, and social issues. These discussions offer an observable window into broader social behavior, making them a useful setting for testing whether LLM agents can reproduce not only fluent text, but also the distributional patterns and interaction dynamics of real online communities. We introduce MiroBench, a benchmark for Reddit discussion simulation built from 4,292 real Reddit threads. MiroBench uses statistical tests to compare generated and real discussions across four major aspects: repetition and semantic uniformity, narrative content, toxicity and aggression, and structural complexity. Experiments across five domains and five models show that current simulators remain distributionally mismatched with real Reddit threads, while a lightweight prompt-based improvement procedure provides only limited gains. MiroBench offers a concrete benchmark for measuring, diagnosing, and improving realism in LLM-based social simulation.

17.
arXiv (CS.CL) 2026-06-12

MentalMARBERT: Domain-Adaptive Pre-training and Two-Stage Fine-Tuning for Arabic Mental Health Disorders Detection

Detecting mental health disorders from Arabic social media text remains challenging due to dialectal variation, informal language, limited high-quality annotated resources, and severe class imbalance. While English mental health natural language processing (NLP) has progressed substantially, Arabic multi-class disorder classification remains insufficiently studied. This study proposes a two-phase framework for Arabic mental health text classification. In phase 1, three Arabic pre-trained language models, AraBERT, CAMeLBERT, and MARBERT, undergo Domain-Adaptive and Task-Adaptive Pretraining (DAPT and TAPT) using a large-scale corpus of unlabeled Arabic mental health tweets. The adapted models are evaluated under a unified protocol to identify the most effective backbone model. In phase 2, the selected model is assessed across four configurations combining single-stage and hierarchical two-stage classification architectures with full fine-tuning and Low-Rank Adaptation (LoRA). To support this study, we constructed a novel annotated Arabic mental health dataset comprising 50,670 tweets across six categories, with strong inter annotator agreement (Krippendorff's Alpha = 0.733, average pairwise agreement = 0.797). Experimental results show that the domain-adapted MARBERT (MentalMARBERT) achieves statistically significant improvements over baseline models in both accuracy and macro-F1. The hierarchical two-stage architecture combined with full fine-tuning achieves the best overall performance, reaching a macro-F1 of 0.861 and an accuracy of 0.877. These findings demonstrate the effectiveness of domain-specific adaptive pretraining and hierarchical classification for Arabic mental health disorder detection.

18.
arXiv (CS.LG) 2026-06-12

Data-driven Lake Water Quality Forecasting for Time Series with Missing Data using Machine Learning

arXiv:2601.15503v2 Announce Type: replace Abstract: Volunteer-led lake monitoring yields irregular, seasonal time series with many gaps arising from ice cover, weather-related access constraints, and occasional human errors, complicating forecasting and early warning of harmful algal blooms. We study Secchi Disk Depth (SDD) forecasting on a 30-lake, data-rich subset drawn from three decades of in-situ records collected across Maine lakes. Missingness is handled via Multiple Imputation by Chained Equations (MICE), and we evaluate performance with a normalized Mean Absolute Error (nMAE) metric for cross-lake comparability. Among six candidates, ridge regression provides the best mean test performance. Using ridge regression, we then quantify the minimal sample size, showing that under a backward, recent-history protocol, the model reaches within 5% of full-history accuracy with approximately 176 training samples per lake on average. We also identify a minimal feature set, where a compact four-feature subset matches the thirteen-feature baseline within the same 5% tolerance. Bringing these results together, we introduce a joint feasibility function that identifies the minimal training history and fewest predictors sufficient to achieve the target of staying within 5% of the complete-history, full-feature baseline. In our study, meeting the 5% accuracy target required about 64 recent samples and just one predictor per lake, highlighting the practicality of targeted monitoring. Hence, our joint feasibility strategy unifies recent-history length and feature choice under a fixed accuracy target, yielding a simple, efficient rule for setting sampling effort and measurement priorities for lake researchers.

19.
bioRxiv (Bioinfo) 2026-06-16

DMcloud: Macromolecular Structure Modeling Using Local Structure Fitting for Medium to Low Resolution cryo-EM maps

Cryogenic electron microscopy (cryo-EM) has become an essential experimental approach in structural biology for determining macromolecular structures. When the resolution of a cryo-EM map is worse than approximately 5[A], fitting known or predicted molecular models into the map becomes a common strategy for interpretation. However, accurately fitting biomolecular models into cryo-EM maps, particularly for large macromolecular complexes, remains challenging when the input structure models contain errors or are in a conformation different from that represented in the map. Here, we present DMcloud, a method for local structure fitting of proteins and nucleic acids in cryo-EM maps. Instead of forcing an entire input model into the map, DMcloud divides input structures into local regions, identifies regions that are supported by the density, removes unsupported regions, and assembles the retained regions into a final model. We benchmarked DMcloud on 176 cryo-EM maps, including intermediate and high-resolution maps that include proteins, DNAs, or RNAs. For EM maps in the 5.0-10.0 [A] and 2.5-5.0 [A] resolution ranges, DMcloud achieved average sequence modeling coverage of 0.49 and 0.70, respectively. For DNA/RNA maps, DMcloud achieved an average sequence coverage of 0.75. Across all datasets, DMcloud consistently outperformed existing methods in model accuracy, map-model correlation, and modeling coverage.

20.
arXiv (CS.LG) 2026-06-19

Approximate Next Policy Sampling: Replacing Conservative Target Policy Updates in Deep RL

arXiv:2605.05481v2 Announce Type: replace Abstract: We revisit a classic "chicken-and-egg" problem in reinforcement learning: to safely improve a policy, the value function must be accurate on the state-visitation distribution of the updated policy. That distribution over states is unknown and cannot be sampled for the purposes of training the value function. Conservative updates solve this problem, but at the cost of shrinking the policy update. This paper explores an alternative solution, Approximate Next Policy Sampling (ANPS), which addresses the problem by modifying the training distribution rather than constraining the policy update. ANPS is satisfied if the distribution of the training data approximates that of the next policy. To demonstrate the feasibility and efficacy of ANPS, we introduce Stable Value Approximate Policy Iteration (SV-API). SV-API modifies the standard approximate policy iteration loop to hold the target policy fixed while an iteratively updated behavioral policy gathers relevant experience. It only commits to a new policy once a convergence criterion has been met. If certain stability criteria are met, the update is guaranteed to be safe; otherwise, it remains no less safe than standard approximate policy iteration. Applying SV-API to PPO yields Stable Value PPO (SV-PPO), which matches or improves performance on high-dimensional discrete (Atari) and continuous control benchmarks while executing substantially larger target policy updates. These results demonstrate the viability of ANPS as a new solution to this classic challenge in RL.

21.
arXiv (CS.CL) 2026-06-16

SHARD: Safe and Helpful Alignment via Self-Reframing Distillation

Large language models often struggle with sensitive prompts. They may refuse outright, provide generic safety boilerplate, or fail to address the user's legitimate informational needs that can be answered safely. We introduce SHARD, a self-reframing distillation method to improve safe-helpfulness. It first rewrites sensitive prompts to surface benign intent using philosophical guidelines, then reframes its original responses into safe, more helpful ones, and finally fine-tunes the model on its self-reframed responses. Across DNA and the English subset of LINGUASAFE, SHARD improves helpfulness for most model families while preserving safety. It also remains competitive with distillation from a larger teacher model, suggesting that models can internalize safe and helpful behavior elicited from their own. Warning: This paper contains content that may be offensive or harmful.

22.
arXiv (CS.CL) 2026-06-11

ASRU: Activation Steering Meets Reinforcement Unlearning for Multimodal Large Language Models

Multimodal large language models (MLLMs) may memorize sensitive cross-modal information during pretraining, making machine unlearning (MU) crucial. Existing methods typically evaluate unlearning effectiveness based on output deviations, while overlooking the generation quality after unlearning. This can easily lead to hallucinated or rigid responses, thereby affecting the usability and safety of the unlearned model. To address this issue, we propose ASRU, a controllable multimodal unlearning framework that incorporates generation quality as a core evaluation objective. ASRU first induces initial refusal behavior through activation redirection, and then optimizes fine-grained refusal boundaries using a customized reward function, thereby achieving a better trade-off between target knowledge unlearning and model utility. Experiments on Qwen3-VL show that ASRU significantly improves unlearning effectiveness (+24.6%) on average and generation quality (5.8X) on average while effectively preserving model utility, using only a small amount of retained supervision data.

23.
arXiv (CS.CV) 2026-06-16

Random Erasing vs. Model Inversion: A Promising Defense or a False Hope?

Model Inversion (MI) attacks pose a significant privacy threat by reconstructing private training data from machine learning models. While existing defenses primarily concentrate on model-centric approaches, the impact of data on MI robustness remains largely unexplored. In this work, we explore Random Erasing (RE), a technique traditionally used for improving model generalization under occlusion, and uncover its surprising effectiveness as a defense against MI attacks. Specifically, our novel feature space analysis shows that models trained with RE-images introduce a significant discrepancy between the features of MI-reconstructed images and those of the private data. At the same time, features of private images remain distinct from other classes and well-separated from different classification regions. These effects collectively degrade MI reconstruction quality and attack accuracy while maintaining reasonable natural accuracy. Furthermore, we explore two critical properties of RE including Partial Erasure and Random Location. Partial Erasure prevents the model from observing entire objects during training. We find this has a significant impact on MI, which aims to reconstruct the entire objects. Random Location of erasure plays a crucial role in achieving a strong privacy-utility trade-off. Our findings highlight RE as a simple yet effective defense mechanism that can be easily integrated with existing privacy-preserving techniques. Extensive experiments across 37 setups demonstrate that our method achieves state-of-the-art (SOTA) performance in the privacy-utility trade-off. The results consistently demonstrate the superiority of our defense over existing methods across different MI attacks, network architectures, and attack configurations. For the first time, we achieve a significant degradation in attack accuracy without a decrease in utility for some configurations.

24.
arXiv (CS.CL) 2026-06-18

DSB: Dynamic Sliding Block Scheduling for Diffusion LLMs

Diffusion large language models (dLLMs) have emerged as a promising alternative for text generation, distinguished by their native support for parallel decoding. In practice, block inference is crucial for avoiding order misalignment in global bidirectional decoding and improving output quality. However, the widely-used fixed, predefined block (naive) schedule is agnostic to semantic difficulty, making it a suboptimal strategy for both quality and efficiency: it can force premature commitments to uncertain positions while delaying easy positions near block boundaries. In this work, we analyze the limitations of naive block scheduling and disclose the importance of dynamically adapting the schedule to semantic difficulty for reliable and efficient inference. Motivated by this, we propose Dynamic Sliding Block (DSB), a training-free block scheduling method that uses a sliding block with a dynamic size to overcome the rigidity of the naive block. To further improve efficiency, we introduce DSB Cache, a training-free KV-cache mechanism tailored to DSB. Extensive experiments across multiple models and benchmarks demonstrate that DSB, together with DSB Cache, consistently improves both generation quality and inference efficiency for dLLMs. Code is released at https://github.com/lizhuo-luo/DSB.

25.
medRxiv (Medicine) 2026-06-15

GLLaucoMed: A Secure LLM-Powered Agentic Workflow for Automated Medication Extraction from Free-Text Glaucoma Clinical Notes

Purpose: To evaluate the efficacy of large language models (LLMs) in extracting medication-related information from glaucoma clinical notes in the electronic health record (EHR). Design: Cross-sectional. Subjects: 1,250 subjects in the Bascom Palmer Ophthalmic Repository. Methods: Extracted clinical notes from glaucoma-related encounters between 2014 and 2024 were labeled by two glaucoma specialists with a third serving as an adjudicator. Graders were asked to label current topical medications (CTM), proposed changes to topical medications ({Delta}TM), current oral medications (COM), and proposed changes to oral medications ({Delta}OM) in a structured fashion. The dataset was split into development (10%), validation (10%), and test (80%) sets stratified by clinician. Development and validation sets were used to engineer and refine prompts, and the held-out test set was used for model assessment. Five LLMs (Claude Opus 4.6, DeepSeek-V3.2, GPT 5.2, Grok 4.1, and Qwen3.6-35B-A3B) were accessed via Microsoft Azure AI Foundry within a HIPAA-compliant environment. Inter-grader agreement was assessed with Gwet AC1. LLM performance was initially assessed in a binary fashion with F1 scores, and the degree of text match among positive cases was evaluated using exact match accuracy and Jaccard Index (JI). Main Outcome Measures: F1 score, exact match accuracy, JI. Results: Gwet AC1 for intergrader agreement was 0.799, 0.888, 0.985, and 0.988 for CTM, {Delta}TM, COM, and {Delta}OM, respectively. F1 scores for CTM were 0.985, 0.971, 0.978, 0.968, and 0.970 for Claude, Deepseek, GPT, Grok, and Qwen, respectively; for {Delta}TM: 0.905, 0.826, 0.897, 0.842, 0.855, respectively; for COM: 0.923, 0.887, 0.899, 0.906, 0.894, respectively; for {Delta}OM: 0.958, 0.815, 0.937, 0.835, 0.940, respectively. Among positive cases, range of exact match accuracies for CTM (N=1354) was 0.730- 0.882 and range of JIs was 0.809-0.918. For {Delta}TM (N=404), exact match accuracy range was 0.619-0.780 and JI range was 0.668-0.827. For COM (N=47), exact match accuracy range was 0.766-0.872 and JI range was 0.765-0.870. For {Delta}OM (N=25), exact match accuracy range was 0.583-0.920 and JI range was 0.583-0.922. Conclusions: The GLLaucoMed pipeline demonstrated high performance in extracting and standardizing medication data from unstructured clinical notes, including both current medications and proposed changes. Claude and GPT exhibited the strongest performance.